NSF proposals get judged on two criteria: intellectual merit and broader impacts. The second one is where a lab’s public presence stops being optional and starts being evidence.
What reviewers are weighing
A broader impacts section makes claims: public engagement, mentoring, outreach to underrepresented groups, training the next generation of scientists. A reviewer reading those claims has no way to verify them from the proposal text alone. What they can do is check whether the lab’s public presence backs the claims up, and an experienced reviewer often does exactly that.
A section describing an active mentoring program, next to a website with no mention of it anywhere, reads as a claim without support. The gap doesn’t need to be dishonest to hurt the proposal. A reviewer noticing it is enough.
What good documentation looks like
Specific and dated beats general and vague. “The lab has hosted five high school interns since 2023, listed here with the projects they worked on” does more than a sentence claiming a commitment to STEM education. A page or a handful of social posts documenting the actual events, with photos and names where appropriate, gives a reviewer something concrete to point to when they write their own assessment.
This works both directions. A well-documented outreach history is easier to write into next year’s proposal, since the evidence already exists instead of needing to be reconstructed from memory during proposal season.
What weak documentation looks like, by comparison
The most common failure isn’t dishonesty. It’s a proposal that describes a real, ongoing mentoring relationship with a local high school in careful detail, next to a website that has never once mentioned it. The activity is real. The documentation isn’t. A reviewer has no way to distinguish that gap from a claim that was never followed through on, and the benefit of the doubt doesn’t always go the lab’s way.
Another common version: a broader impacts section that lists activities from three years ago as if they’re current, because nobody updated the language after the program quietly ended. Reviewers who follow up on specifics, which some do, notice this quickly.
Where this lives
Most of this doesn’t belong buried in a PDF. A dedicated outreach or broader impacts page on the lab website, updated as events happen rather than assembled once a year under deadline, does the job. Social posts from the events themselves, at the time they happen, add a second layer of evidence with a timestamp a reviewer can check.
Building the habit into the lab calendar
The labs that do this well treat documentation as part of the event itself, not a follow-up task. A photo taken during the high school visit, posted or filed the same week, takes less time than trying to reconstruct what happened and when six months later. Assigning this to whoever organizes the event, as a standard last step, keeps it from depending on someone remembering separately.
This isn’t only an NSF question
NSF names broader impacts explicitly as a review criterion, which is why it comes up first, but the underlying expectation shows up elsewhere too. NIH reviewers weigh a project’s significance and its training or mentoring plan. Private foundations and industry-adjacent funders often ask directly about public engagement or workforce development in their own applications. The specific language changes by funder. The need for a documented, checkable public record of the activity a proposal describes doesn’t.
A lab that builds the habit of documenting outreach for one funder’s requirements ends up with material that transfers cleanly to every other funder’s version of the same question, which makes the upfront habit worth more than it looks the first time.
The renewal problem this solves
More programs now ask what happened to the broader impacts promised in the last proposal, on top of what’s promised in this one. A lab that documented outreach as it happened has an easy answer. A lab that promised outreach and never wrote any of it down is reconstructing a history from memory, under a deadline, and it shows.
Common questions
Why does a lab's website matter for broader impacts review?
Because reviewers have no way to verify claims like mentoring or outreach from the proposal text alone. A website or social account documenting those activities, with dates and specifics, gives a reviewer something concrete instead of a promise to take on faith.
What does strong broader impacts documentation look like?
Specific and dated, not general. A page listing five high school interns hosted since a given year, with the projects they worked on, does more than a paragraph describing a commitment to STEM education in the abstract.
When should outreach activities get documented, before or after a proposal is due?
As they happen, not during proposal season. A lab that logs outreach the week it occurs walks into every future proposal with evidence already assembled, instead of reconstructing a history from memory under a deadline.
Do broader impacts claims get checked at renewal time?
Increasingly, yes. More programs ask what happened to the broader impacts promised in the prior proposal, not only what's promised in the new one, which makes a documented history more valuable than it used to be.